That is the sort of explanation I expected, but when I read the article, it presents a different causal sequence, in which photosynthesizing plants become ineffective, not because of increasing temperatures, but because of decreasing atmospheric CO2, as a result of its photodissociation. [1]

As presented in the news article, there is no explanation for why the apparently still-abundant oxygen does not react with, for example, the increased methane flux that the article also mentions, or just the carbon that has been liberated by dissociation, to maintain a CO2 level adequate for photosynthesis.

I don't doubt that the researchers have answers to these questions (quite likely the one you gave) but something seems to have gone missing in the creation of this news article.

[1] The article actually says "due to the gas absorbing the heat and breaking down", but AFAIK, CO2 is pretty resistant to this. Perhaps this is a garbled version of CO2 reacting more vigorously with rocks and soil at higher temperatures?

Oh, I see. Yeah, well my understanding of this is Earth as a self-balancing system:

The Sun gets brighter over time. Plants (over tens of thousands of years) absorb enough CO2 to keep the temperature from rising too much. But eventually the Sun is so bright that in order to keep the temperature in the right range, plants would have to absorb so much CO2 that there wouldn’t be enough for effective photosynthesis so that eventually in the future this negative feedback mechanism would no longer be able to compensate for the increased brightness of the Sun and so the temperatures would rise, perhaps eventually killing off enough plants to allow CO2 to start to be released and then you get runaway greenhouse effect like Venus.

That is all ignoring humans, of course.

EDIT: actually, it may make more sense to see the feedback coming from the other side of the temperature range: plants will keep absorbing as much CO2 as they can until temperatures drop (think ice ages) and photosynthesis slows. Volcanic emissions (and also occasional natural emissions of fossil carbon) keep a certain amount of CO2 being added which is then kept in equilibrium by plants. As the Sun warms, it takes a lower and lower amount of CO2 to get to ice ages that slow photosynthesis. The Sun over the ages increases in brightness until eventually CO2 levels can’t drop any more without slowing photosynthesis to a stop, so you stop getting ice ages and then eventually the temperatures get so high that photosynthesis is limited by die-off from high temperatures, removing the negative feedback mechanism of photosynthesis and you get runaway greenhouse effect.

That’s why pre-industrial CO2 levels were so low. Primates developed and later intelligent humans developed during the last few million and few hundred thousand years, when CO2 levels have been at their lowest levels ever in our planet’s history (and it shows that we evolved under low CO2 conditions, since human mental processing becomes somewhat less effective at even moderate CO2 levels like 800-1000ppm). Plants (unlike human brains) actually like higher CO2 levels and so they do function as an effective negative feedback mechanism as long you don’t outrun their photosynthesis capability and as long as you don’t kill them off with higher temperatures.

That is in line with what I thought was the scientific consensus too, but when I read the abstract (I should have done that first!) I see it saying "Our results suggest that the planetary carbonate–silicate cycle will tend to lead to terminally CO2-limited biospheres and rapid atmospheric deoxygenation", where I assume the carbonate-silicate cycle is the reactions between CO2 and rocks/soil, and if so, then this model seems to be suggesting that this will come into play before a runaway greenhouse scenario.

In that case, I think they’re saying that both CO2 and O2 will be absorbed over time by minerals (as well as CO2 being absorbed by plants), and once CO2 levels drop too much, photosynthesis stops (or slows), which means oxygen stops (or slows) being replenished as well, so oxidization of minerals will also lead to a reduction in oxygen levels.

(And less volcanic activity over time also means less CO2 being added back to the atmosphere, so the chemical weathering process that absorbs both CO2 and O2 into minerals will no longer have an active counterbalancing force from volcanoes.)